Exploiting the potential anti-tumor effect of stem cell grafts remains an attractive concept. One strategy for enhancing graft vs. tumor reactions without aggravating graft vs. host disease (GVHD) would be to selectively target an immune response against a defined tumor specific antigen. This could be accomplished by eliciting an anti-tumor immune response in stem cell transplant (SCT) donors by active immunization, prior to adoptive transfer to the recipient. Previous work in human SCT, demonstrating that the transfer of humoral, and to a lesser extent cellular, antigen-specific immunity to clinically important viral antigens from immune donors to recipients can occur, should favor efforts to transfer donor tumor antigen-specific immunity. In addition, the minimal residual disease burden achievable in many patients by high dose conditioning regimens provides an optimal clinical setting to test the additive effect of adoptive immunotherapy. Finally, and perhaps most compelling, one would predict that immunization of a healthy donor against a tumor antigen would be more easily accomplished, compared with the patient, who may be immunocompromised from the underlying disease or cytotoxic treatment, or both. However, safety considerations would mandate that the tumor vaccine be highly refined, consisting of a well-defined antigen, and that the targeted antigen be truly tumor-specific. Based on our preclinical studies in two murine model systems which provided the rationale for this approach, we pioneered this strategy in the disease indication of multiple myeloma. With FDA approval, the HLA-matched brother of a woman with refractory myeloma (M-protein 3.9 g/dl, 50% marrow plasmacytosis) received two s.c. immunizations of myeloma Id-KLH at one week intervals before marrow harvest (Lancet 345:1016-20, 1995). The demonstration of transfer of T-cell immunity was provided by the recovery of a CD4+, donor-derived, idiotype-specific T-cell line generated from the recipient's blood. Building on the encouraging results from the single patient studied above, a clinical trial of marrow donor immunization in multiple myeloma was opened under an approved FDA IND in collaboration with the University of Arkansas (as the clinical transplantation site). This and future clinical protocols have been designed to explore whether a booster immunization of the recipient might improve the potency and duration of the transferred idiotype-specific response and will support the two specific subaims below. To date, five additional donor-recipient pairs have been enrolled on this protocol. Coincident with the opening of a dedicated SCT unit in the DETI, a new pilot clinical protocol, based entirely within the DETI, has been developed jointly with Michael Bishop, M.D. to replace the existing protocol. This new protocol incorporates two additions to the strategy of donor vaccination which are designed to further optimize the transfer of T-cell immunity from donor to recipient. First, a non-myeloablative conditioning regimen will be used, associated with somewhat attenuated GVHD prophylaxis, which is a potential barrier to transfer of donor tumor antigen-specific T cells. Second, instead of marrow, the stem cell source will be blood mobilized with G-CSF. Because of the larger proportion of T cells in the allograft, the use of peripheral blood as the transfer element could potentially enhance the transfer of tumor immunity. The longterm goal of this project is to transfer highly enriched populations of idiotype-specific T cells from donor to recipient (e.g. tumor-specific DLI). In future years, methods of expansion of primed, donor T cells in vitro will be explored, as well as priming of donor T cells to idiotype in vitro, the latter as a potential alternative to vaccinating healthy donors. (This project was formerly associated with ID Z01 SC 009397 M.)